JP4247232B2 - Electric power steering device - Google Patents

Description

【Technical field】
[0001]
  The present invention relates to an electric power steering device, and more particularly to a control device thereof.
[Background]
[0002]
  An electric power steering device for a vehicle detects a steering torque generated in a steering shaft by operating a steering handle, calculates a steering assist command value that is a control target value of a motor based on the detection signal, In the feedback control circuit, the difference between the steering assist command value, which is the control target value, and the detected motor current value is obtained as a current control value, and the motor is driven by the current control value to assist the steering force of the steering handle. There is something to do.
[0003]
  In such an electric power steering apparatus, as shown in FIG. 7, four field effect transistors FET1 to FET4 are connected to a bridge to form an H bridge circuit having first and second arms. A motor control circuit is used in which a power source V is connected between the input terminals and the motor M is connected between the output terminals.
[0004]
  The first arm FET1 (or the second arm FET2) of the two pairs of FETs constituting the two opposing arms of the H bridge circuit constituting the motor control circuit is set to a current control value. By driving with a PWM signal (pulse width modulation signal) having a duty ratio D determined based on the above, the magnitude of the motor current is controlled.
[0005]
  Further, based on the sign of the current control value, the second arm FET3 is turned on and the first arm FET4 is turned off (or the second arm FET3 is turned off and the first arm FET4 is turned on). Thus, the rotation direction of the motor M is controlled.
[0006]
  When FET3 is in a conducting state, current flows through FET1, motor M, and FET3, and a positive current flows through motor M. When the second arm FET4 is in a conducting state, a current flows through the FET2, the motor M, and the FET4, and a negative current flows through the motor M. This motor control circuit is widely used because the FETs on the same arm are not driven at the same time, so the possibility that the arms are short-circuited is low and the reliability is high.
[0007]
  FIG. 8 shows the relationship between the motor current I (current actually flowing through the motor and different from the detected value of the motor current) and the duty ratio D of the PWM signal. That is, in the state where the steering handle is operated and steering torque is generated, the relationship between the motor current I and the duty ratio D changes as shown by the line (a) in FIG. The steering assist command value Iref, which is the control target value of the motor current, is calculated based on the detected signal of the current, and the current control value E of the difference between the calculated steering assist command value Iref and the detected value i of the motor current to be fed back is calculated. Is output to the motor drive circuit, the duty ratio D for controlling the semiconductor element of the motor drive circuit takes a certain value, and no particular trouble occurs.
[0008]
  However, when the steering handle returns to the straight traveling position by self-aligning torque after turning the steering handle (hereinafter referred to as “handle return”), the steering current is not generated. Although the steering assist command value Iref, which is the control target value, becomes zero, since a back electromotive force is generated in the motor, the relationship between the motor current I and the duty ratio D is as shown by a line (b) in FIG. When the duty ratio D is near zero, a discontinuous portion x occurs in the relationship between the motor current I and the duty ratio D.
[0009]
  On the other hand, the feedback control circuit tries to calculate the current control value E, but since there is no duty ratio D corresponding to the steering assist command value Iref, as shown by the line (c) in FIG. An oscillating current having an amplitude substantially corresponding to the continuous portion is output as the current control value E. The generation of such an oscillating current becomes a source of noise and also a cause of hindering the stability of feedback control.
[0010]
  In order to solve this problem, the applicant of the present invention has described the semiconductor element of the first arm out of a set of two semiconductor elements constituting two arms facing each other of the H bridge circuit constituting the motor drive circuit. The first duty ratio PWM signal determined based on the current control value is driven, and the second arm semiconductor element is driven by the second duty ratio PWM signal defined by the function of the first duty ratio. , Proposed a configuration to drive each independently. According to this configuration, as shown in FIG. 9, the value of the duty ratio D is zero even when the steering torque is not generated such as when the point p and the point o are connected by a straight line and the steering wheel is returned. In the vicinity, there is no discontinuous portion in the relationship between the motor current I and the duty ratio D, and no oscillating current is output as the current control value E, so noise feedback is not generated and stable feedback control is realized. (See Patent Document 1).
[0011]
  As described above, the semiconductor element of the first arm is driven by the PWM signal having the first duty ratio determined based on the current control value, and the semiconductor element of the second arm is driven by the first duty ratio. In the configuration in which each PWM signal having the second duty ratio defined by the function is driven independently, there is no discontinuity in the relationship between the motor current I and the duty ratio D, noise is not generated, and stability is improved. The However, as apparent from FIG. 9, since the relationship between the motor current I and the duty ratio D can be switched in three stages, it is difficult to eliminate chattering due to the switching, and control noise and vibration due to chattering are generated. Inconvenience occurs. The present invention aims to solve the above problems.
[0012]
[Patent Document 1]
  JP-A-9-39810.
DISCLOSURE OF THE INVENTION
[0013]
  An electric power steering apparatus according to the present invention controls an output of a motor that applies a steering assist force to a steering mechanism based on at least a steering assist command value calculated based on a steering torque signal generated in a steering shaft. , A duty ratio calculating means for calculating a duty ratio D1 and a duty ratio D2 for determining the voltage between the motor terminals based on the steering assist command value, and a first and a second including two semiconductor elements connected in series. A power source is connected between the input terminals of the H bridge circuit composed of the arms of the motor, the motor is connected between the output terminals, and the upper semiconductor element of the first arm of the H bridge circuit is driven by the PWM signal of the duty ratio D1. And a motor for driving the lower semiconductor element of the second arm with the PWM signal having the duty ratio D2. And a dynamic circuit, the duty ratio calculating means,Absolute value Vref of motor terminal voltage command value and absolute value K of back electromotive force of motor _T When the following condition (c) is satisfied between ω and
      | Vref | <| K _T ω | ... (c)
  The duty ratio D1 is calculated by the following formula (a), and the duty ratio D2 is calculated by the following formula (b).
      D1 = Vref2 / V _r (A)
      D2 = {Vref2 + sign (Vref2) (V _r − | K _T ω |)} / V _r
            ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ （B）
        Vref: Motor terminal voltage command value
              Vref2: Linearized motor terminal voltage command value
                      = 1/2 (Vref-K _T ω)
              V _r : Voltage supplied to the motor (battery voltage)
              K _T : Back electromotive force constant of motor
              ω: Motor angular velocity
              sign (Vref2): Sign of linearized motor terminal voltage command value Vref2
An electric power steering device characterized by the above.
[0014]
  The duty ratio calculating means includesAbsolute value Vref of motor terminal voltage command value from which noise has been removed and absolute value K of motor back electromotive force from which noise has been removed _T When the following condition (d) is satisfied between ω and
      | Vref | <| K _T ω | ... (d)
  The duty ratio D1 is calculated by the equation (a), and the duty ratio D2 is calculated by the equation (b).
[0015]
  Then, the duty ratio calculating means is configured to output the absolute value Vref of the voltage command value between the motor terminals and the absolute value K of the back electromotive force of the motor. _T When the following condition (f) is satisfied between ω and
      (| Vref |-| K _T ω |) <-Hys ... (f)
        However, Hys: Hysteresis width characteristic value
  The duty ratio D1 is calculated by the equation (a) and the duty ratio D2 is calculated by the equation (b). When the following condition (g) is satisfied, the previous determination result is maintained.
      −Hys <(| Vref | − | K _T ω |) <Hys ... ・ (g)
It is characterized by.
[0016]
  The motor terminal voltage command value Vref and the motor back electromotive force K _T At least one of ω has been subjected to noise removal processing.
[0017]
  The hysteresis width characteristic value Hys is determined according to the magnitude of noise.
[0018]
  The duty ratio calculating means includesA current drive linearization compensator and a current discontinuity compensator are provided, and the current drive linearization compensator receives the voltage command value Vref between motor terminals as an input and linearizes the voltage command value between motor terminals based on the equation (a). The duty ratio D1 corresponding to Vref2 is calculated, and the current discontinuity compensator calculates the duty ratio D2 based on the equation (b) with the linearized motor terminal voltage command value Vref2 as an input.
BEST MODE FOR CARRYING OUT THE INVENTION
[0019]
  Embodiments of the present invention will be described below. First, the basic concept of the present invention will be described. In the present invention, the non-linear control characteristic between the motor current I and the duty ratio D described above with reference to FIG. 9, that is, the non-linear control characteristic composed of a straight line bent in three stages is further improved. Is intended to have a continuous linear characteristic.
[0020]
  As described above with reference to FIG. 9, the motor control circuit of the electric power steering apparatus includes four field effect transistors FET1 to FET4 connected to the bridge, and includes first and second arms. An H-bridge circuit is formed, and FET1 to FET4 have a first duty ratio D1 (hereinafter referred to as duty) determined based on a current control value E which is a difference between a steering assist command value and a detected value of a motor current to be fed back. D1) and a second duty ratio D2 (hereinafter referred to as duty D2).
[0021]
  FIG. 1 shows the motor bridge voltage Vm and motor current I when the FET 1 is driven with a duty D1 and the FET 3 is turned on, that is, with a duty D2 = 100%, and the FET 2 and the FET 4 are turned off. When the motor angular speed ω is (ω = α) and the motor terminal voltage Vm is increased from the minus side to the plus side, the motor current i is increased when the motor terminal voltage Vm is (Vm = 0). Suddenly becomes zero (i = 0). When the motor angular velocity ω is (ω = −α) and the motor terminal voltage Vm is decreased from the plus side to the minus side, the motor current i suddenly becomes zero when the motor terminal voltage Vm is (Vm = 0). (I = 0).
[0022]
  In the above description, the motor terminal voltage Vm is described. However, since the duty ratio D is a ratio that determines the motor terminal voltage Vm, the motor terminal voltage can be replaced with the duty ratio. Further, if the combination of FET1 and FET3 is a combination of FET2 and FET4, the rotational direction of the motor is reversed, but the operation is essentially the same. Therefore, in the following description, FET1 and FET3 will be described.
[0023]
  In order to improve the nonlinear characteristics described above, in the present invention, in the H bridge circuit having the first and second arms, FET 1 is driven by duty D1, and FET 3 is driven by duty D2. The duty D1 is set by the following formula (a), and the duty D2 is set by the following formula (b).
[0024]
    D1 = Vref2 / V_r・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ (A)
    D2 = {Vref2 + sign (Vref2) (V_r− | K_Tω |)} / V_r(B)
        However, Vref2 = 1/2 (Vref−K_Tω)
              | Vref | <| K_TRange of ω |
        Where Vref: Motor terminal voltage command value
                Vref2: Linearized motor terminal voltage command value
                V_r: Voltage supplied to the H-bridge circuit (battery voltage)
                K_T: Back electromotive force constant of motor
                ω: Motor angular velocity
                sign (Vref2): Linearized motor terminal voltage command value Vref2
                Sign.
[0025]
  Hereinafter, a method for calculating the above-described duty D1 and duty D2 will be described.
[0026]
  The basic equation for PWM signal driving of the H-bridge circuit is expressed by the following equation (1).
[0027]
    Vm = (D1 + D2) Vr-sign (D1) Vr-K_Tω (1)
        Where Vm: Voltage between motor terminals
          D1: Upper stage duty for driving the upper stage FET (value -1 to +1)
          D2: Lower stage duty (value -1 to +1) for driving the lower stage FET
          V_r: Voltage supplied to the H-bridge circuit (battery voltage)
          K_T: Back electromotive force constant of motor
          ω: Motor angular velocity
  Normally, the duty D2 is fixed to 100% (D2 = 1.0), and only the duty D1 is changed. Therefore, for example, the voltage Vm between the motor terminals when 30% of the battery voltage (D1 = 0.3) is applied to the motor is obtained from the equation (1) because the sign (0.3) of D1 is positive. It becomes as follows.
[0028]
    Vm = (0.3 + 1) V_r-Sign (0.3) V_r-K_Tω
        = 0.3V_r-K_Tω
  However, in the prior art described above (see Patent Document 1), the duty D2 is calculated by the following equation (2) in order to solve the inconvenience as described above.
[0029]
    D2 = D1 + sign (D1) x B (2)
      Where B is a constant
  Then, the constant B is determined so that the relationship between the duty D1 and the motor current I becomes the characteristic shown in FIG. Since the internal resistance of the motor can be treated as a constant value, the characteristic diagram shown in FIG. 9 is established even if the motor current I is replaced with the motor terminal voltage Vm.
[0030]
  Hereinafter, the determination of the constant B will be described. In equation (1), duty D1 and motor back electromotive force K_TWhen the condition that the sign is different from ω is included, the expression (1) can be rewritten into the following expression (3).
[0031]
    Vm = (D1 + D2) Vr-sign (D1) Vr + sign (D1) | K_Tω | ...
          .... (3)
  Equation (3) represents the dead band characteristics shown in FIG. If the condition that the motor terminal voltage Vm is zero (Vm = 0) when the upper duty D1 is zero (D1 = 0) is substituted into the equation (3), the constant B can be expressed by the following equation (4). it can.
[0032]
    0 = (0 + D2) Vr-sign (0) Vr + sign (0) | K_Tω |
      = D2 Vr -Vr + | K_Tω | and substituting equation (2) into this,
    0 = {D1 + sign (D1) * B} Vr-Vr + | K_Tω |
      = BVr -Vr + | K_Tω |
    B = 1- {| K_Tω | / Vr} (4)
  That is, since the constant B is determined by the above equation (4), the duty D2 expressed by the equation (2) is a function of the duty D1.
[0033]
  FIG. 2 is a rewrite of FIG. 9 described above, and illustrates the improvement of discontinuous characteristics between the motor terminal voltage Vm and the motor current I shown in FIG. The duty D1 is shown instead of Vm. Further, as described above, since the motor internal resistance can be treated as a constant value on the vertical axis, the motor current is replaced with the voltage between the motor terminals.
[0034]
  FIG. 2 shows that the discontinuous characteristic is converted into the continuous characteristic by converting the part A1 'showing the discontinuous characteristic into the part A1 and the part A2' into the part A2.
[0035]
  The characteristic equation of the portion A1 'and the portion A2' can be expressed by the following equation (5) if the duty of this portion is D1 '.
[0036]
    Vm = Vr D1 '-K_Tω (5)
  If the duty can be defined by D1 ′ and the duty by D1, the discontinuous characteristic can be converted into a continuous characteristic. Expression (2), Expression (4), and Expression (5) are substituted into Expression (1). First, equation (5) is substituted into equation (1).
[0037]
    Vm = (D1 + D2) Vr-sign (D1) Vr-K_Tω
    Vr D1 '-K_Tω = (D1 + D2) Vr-sign (D1) Vr-K_Tω
    Vr D1 '= (D1 + D2) Vr-sign (D1) Vr
Substituting equation (2) into D2 of this equation
    Vr D1 '= {D1 + (D1 + sign (D1) × B)} Vr
                  -Sign (D1) Vr
    D1 '= 2D1 + sign (D1) (B-1)
Solving this equation with D1
    D1 = 1/2 {D1'-sign (D1) (B-1)}
Substituting equation (4) into B in this equation
    D1 = 1/2 {D1'-sign (D1) {| K_Tω | / Vr}
Upper duty D1 and K_TIf we add the condition that ω is a different sign,
    D1 = 1/2 {D1 '-(K_Tω / Vr)} (6)
Since sign D1 can be deleted from the right side of equation (6) and the absolute value can be removed, duty D1 'can be defined by duty D1.
[0038]
  In the above explanation, the discontinuous characteristic can be converted into the continuous characteristic by converting the part A1 ′ of the discontinuous characteristic between the motor terminal voltage Vm and the motor current I in FIG. However, in FIG. 2, the characteristics of the voltage Vm between the motor terminals and the duty D1 are continuous characteristics that are bent in three stages of po-q, and are therefore bent in these three stages. The continuous characteristic p-o-q is converted into a completely linear continuous characteristic p-q as shown in FIG.
[0039]
  In this embodiment, the voltage between the motor terminals is controlled by calculating the voltage command value Vref between the motor terminals from the difference between the current command value Iref for controlling the motor current and the detected motor current i. Since the ratio value is calculated and determined as a voltage value, in the following description, the motor terminal voltage command value Vref will be described.
[0040]
  First, the motor terminal voltage command value Vref is mapped to the linearized motor terminal voltage command value Vref2 which is the second voltage command value according to the equation (6). Here, the meaning of “mapping” is a voltage command between motor terminals so that the continuous characteristic p-oq bent in three stages shown in FIG. 2 is converted into a completely linear continuous characteristic pq shown in FIG. This means that the value Vref is converted into the linearized motor terminal voltage command value Vref2.
[0041]
  The mapping process is a process in which duty D1 = Vref2 / Vr and duty D1 '= Vref / Vr are set, and line A1 shown in FIG. 2 is converted to A1' and line A2 is converted to A2 '. This conversion corresponds to the range of pq in FIG.
    Nonlinear characteristics | Vref | <| K_TThe range of ω |.
[0042]
  By substituting D1 = Vref2 / Vr and D1 '= Vref / Vr into the equation (6), the equation (6) is expressed by the following equation (7), and the mapping process is performed by the equation (7). Vref2 / Vr = 1/2 {(Vref / Vr)-(K_Tω / Vr)}
    Vref2 = 1/2 (Vref-K_Tω) ... (7)
  The calculation of duty D1 will be described. In the mapping process described above, the duty D1 is D1 = Vref2 / Vr, and Vref2 is expressed by the above equation (7). Therefore, the duty D1 is expressed by the following equation (a).
[0043]
    D1 = {1/2 (Vref-K_Tω)} / Vr
        = Vref2 / Vr (a)
  In an actual control circuit to be described later, compensation processing such as dead time compensation and duty dither addition processing is performed on the duty D1 represented by the equation (a). Whether or not this processing is performed is determined. It is an optional choice. The duty D1 determined by the above equation (a) does not include the results of compensation processing such as dead time compensation and duty dither addition processing.
[0044]
  Next, the calculation of duty D2 will be described. By substituting Equations (4) and (7) into Equation (2), the duty D2 can be expressed by Equation (b) below.
[0045]
    D2 = D1 + sign (D1) x B (2)
        = (Vref2 / Vr) + sign (Vref2 / Vr)
          × {1- (| K_Tω | / Vr)}
          = {Vref2 + sign (Vref2) (V_r− | K_Tω |)} / V_r...
            ... (b)
  That is, the duty D2 can be expressed by an expression that does not include the duty D1 obtained by the equation (a). This means that the duty D2 can be determined independently of the duty D1. .
[0046]
  The mapping process described above is executed in the range of p-q in the characteristic diagram shown in FIG. 2, and the absolute value Vref of the voltage command value between the motor terminals and the absolute value K of the motor back electromotive force._TThe following condition (c) is required to be satisfied with ω.
[0047]
          | Vref | <| K_Tω ｜ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ (c)
  When the condition (c) is satisfied, the duty D1 is calculated by the equation (a), and the duty D2 is calculated by the equation (b). When the condition is not satisfied, the duties D1 and D2 are calculated by a normal method without executing the mapping process.
[0048]
  However, in the vicinity of the limit of the condition, that is, in the vicinity of the point q in FIG. 2, the values of the duties D1 and D2 obtained by the mapping process greatly differ depending on whether or not this condition is satisfied. That is, when the condition is satisfied, the mapping process from the point o to the point q is performed and converted into the completely linear continuous characteristic pq shown in FIG. 3, but when the condition is not satisfied, the mapping process is not performed and the normal method Thus, the duties D1 and D2 are calculated. In this case, the characteristic diagram shown in FIG. 2 remains as a continuous characteristic p-o-q bent in three stages.
[0049]
  Motor terminal voltage command value Vref, motor back electromotive force K_TWhen noise is included in ω, the condition is satisfied or not satisfied (the condition is satisfied / not satisfied) near the limit of the condition (near the q point in FIG. 2), and the above mapping process is executed / not satisfied. Since the execution is frequently switched, the values of the duties D1 and D2 frequently fluctuate, chattering occurs, and noise and vibration are generated.
[0050]
  Therefore, in the above mapping process, the voltage command value Vref between the motor terminals and the motor back electromotive force K are used using a filter or the like._TAfter the noise component is removed from ω, whether the above condition is satisfied or not is determined.
[0051]
  That is, the absolute value Vref of the motor terminal voltage command value from which noise is removed and the absolute value K of the motor back electromotive force from which noise is removed_TWhen the following condition (d) is satisfied between ω and
          | Vref | <| K_Tω ｜ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ (d)
It is preferable that the duty D1 for executing the mapping process is calculated by the equation (a) and the duty D2 is calculated by the equation (b). When the condition is not satisfied, the duties D1 and D2 are calculated by a normal method without executing the mapping process.
[0052]
  In order to prevent chattering from occurring in the mapping process, a hysteresis characteristic may be added to the above condition.
[0053]
  In order to prevent chattering from occurring near the limit value of the hysteresis characteristic, the previous determination result may be maintained regardless of whether the previous condition is satisfied or not.
[0054]
  That is, the absolute value Vref of the motor terminal voltage command value from which noise is removed and the absolute value K of the motor back electromotive force from which noise is removed_TWhen the following condition (f) is satisfied between ω and
          (| Vref |-| K_Tω |) <-Hys ... (f)
              However, Hys: Hysteresis width characteristic value
The duty D1 for executing the mapping process may be calculated by the equation (a), and the duty D2 may be calculated by the equation (b).
[0055]
  At this time, the condition (f) is not satisfied (Hys <(| Vref | − | K_Tω |)), it is assumed that the duties D1 and D2 are calculated by a normal method without executing the mapping process.
[0056]
  Furthermore, when the following condition (g) is satisfied, the determination result of the previous condition (satisfaction or non-satisfaction of the condition (f)) may be maintained without executing the mapping process.
[0057]
          −Hys <(| Vref | − | K_Tω |) <Hys (g)
  The hysteresis width characteristic value Hys is the motor terminal voltage command value Vref and the motor back electromotive force K._TIt may be appropriately determined by experiment or the like according to the magnitude of noise included in ω.
[0058]
  Next, an outline of an electric power steering apparatus suitable for carrying out the present invention will be described with reference to FIGS. FIG. 4 is a diagram for explaining the outline of the configuration of the electric power steering apparatus. The shaft 2 of the steering handle 1 is coupled to a steering wheel tyrod 8 through a reduction gear 4, universal joints 5a and 5b, and a pinion rack mechanism 7. ing. A torque sensor 3 that detects the steering torque of the steering handle 1 is provided on the shaft 2, and a motor 10 that assists the steering force is coupled to the shaft 2 via a clutch 9 and a reduction gear 4.
[0059]
  The electronic control circuit 13 that controls the power steering device is supplied with electric power from the battery 14 via the ignition key 11. The electronic control circuit 13 calculates a steering assist command value based on the steering torque detected by the torque sensor 3 and the vehicle speed detected by the vehicle speed sensor 12, and supplies the steering assist command value to the motor 10 based on the calculated steering assist command value. To control the current.
[0060]
  The clutch 9 is controlled by the electronic control circuit 13. The clutch 9 is coupled in a normal operation state, and is disconnected when the electronic control circuit 13 determines that the power steering apparatus is out of order and when the power is turned off.
[0061]
  FIG. 5 is a block diagram of the electronic control circuit 13. In this embodiment, the electronic control circuit 13 is mainly composed of a CPU. Here, functions executed by a program in the CPU are shown. For example, the phase compensator 21 does not indicate the phase compensator 21 as independent hardware, but indicates a phase compensation function executed by the CPU.
[0062]
  Hereinafter, functions and operations of the electronic control circuit 13 will be described. The steering torque signal input from the torque sensor 3 is phase compensated by the phase compensator 21 in order to increase the stability of the steering system, and is input to the steering assist command value calculator 22A. A vehicle speed signal detected by the vehicle speed sensor 12 is also input to the steering assist command value calculator 22A.
[0063]
  The steering assist command value calculator 22A calculates a steering assist command value (current command value) Iref by a predetermined formula based on the input steering torque signal, vehicle speed signal, and detected motor current value i. The current controller 22B calculates a motor terminal voltage command value Vref based on the input steering assist command value (current command value) Iref and the detected motor current value i.
[0064]
  The duty ratio calculation device 30 constituting the duty ratio calculation means includes a current drive linearization compensator 23, a current discontinuity compensator 24, and a compensation adder 25. The compensation adder 25 includes a multiplier 26 and dead time compensation. This is a calculation means that outputs a duty D1, a duty D2, and a motor drive direction signal.
[0065]
  The current drive linearization compensator 23 receives the motor terminal voltage command value Vref, the battery voltage Vr, and the motor angular speed ω (detected by a motor angular speed sensor (not shown) or estimated from the motor terminal voltage and motor current) as inputs. The linearized motor terminal voltage command value Vref2 is calculated based on the equations (6) and (7). The calculated value Vref2 is input to the current discontinuity compensator 24 and the compensation adder 25.
[0066]
  The compensation adder 25 calculates the duty D1 based on the above equation (a). The multiplier 26 multiplies the linearized motor terminal voltage command value Vref2 by a predetermined gain K, and the dead time compensator 27, Compensation processing such as dead time compensation and duty dither addition processing is performed in the duty dither adder 28, and the compensated duty D1 is calculated.
[0067]
  The current discontinuity compensator 24 calculates the duty D2 based on the equation (b), and calculates the duty D2 from the linearized motor terminal voltage command value Vref2.
[0068]
  The calculated duty D1 and duty D2 and the motor drive direction signal output from the current drive linearization compensator 23 are input to the motor drive circuit 35.
[0069]
  FIG. 6 shows an example of the configuration of the motor drive circuit 35. The motor drive circuit 35 comprises an FET gate drive circuit 36 and an H bridge circuit 37 comprising FET1 to FET4, and drives the FET1 to FET4 based on the inputted upper stage duty D1 and lower stage duty D2 and the motor drive direction signal.
[0070]
  The motor current detection circuit 38 detects the magnitude of the positive current based on the voltage drop across the resistor R1, and detects the magnitude of the negative current based on the voltage drop across the resistor R2. The detected motor current value i is fed back to the steering assist command value calculator 22A and the current controller 22B.
[0071]
  Here, dead time compensation and duty dither addition processing will be described. First, dead time compensation will be described. In the motor drive circuit using the H bridge circuit, when the signal is switched from H to L based on the duty D of the PWM signal or when the signal is switched from L to H, the two arms of the H bridge circuit are simultaneously turned on. In order to prevent a short circuit, a dead time is provided at the switching point of the PWM signal. Since dead time compensation is not the subject of the present invention, a description thereof is omitted here, but it is described in Japanese Patent Application Laid-Open No. Hei 8-142844 filed by the present applicant.
[0072]
  Next, the duty dither addition process will be described. In the motor drive circuit using the H-bridge circuit, when the duty D of the PWM signal is near zero, a dead zone is generated in the duty D vs. motor current characteristic, the control response is poor, and a natural steering feeling cannot be obtained. Therefore, in the vicinity of the dead zone, a voltage dither signal is supplied to the motor to improve control responsiveness and bring it closer to a natural steering feeling. Since the duty dither addition processing is not the subject of the present invention, the description thereof is omitted here, but is described in Japanese Patent Application Laid-Open No. 2003-11834, which is filed by the present applicant.
[0073]
  As described above, according to the present invention, since the duty ratio D versus the motor current characteristic shows a linear characteristic when the duty ratio D is near zero, the duty ratio D is the same as the control device of the conventional electric power steering apparatus. In addition to eliminating the discontinuous portion of the duty ratio D vs. motor current characteristics near zero, it is possible to eliminate the gradual continuous characteristics, so even when returning to the straight line after turning the steering wheel The feedback characteristic is not changed, and an extremely smooth steering feeling can be given without giving a sense of incongruity to the steering feeling.
[0074]
  In addition, the back electromotive force generated in the motor when the handle is returned changes continuously, and no stepwise change occurs, so chattering does not occur, noise does not occur due to chattering, and noise is generated in car radio etc. There is a remarkable effect which is not seen in the conventional apparatus, such as there is no fear of causing it.
[Industrial applicability]
[0075]
  BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electric power steering apparatus for a vehicle. In a motor drive circuit using an H bridge circuit in which semiconductor elements are bridge-connected, a duty D vs. motor is generated when the duty D of a PWM signal for driving the semiconductor elements is near zero. The discontinuous characteristic of the current characteristic is a linear characteristic, so that the control response is improved and a natural steering feeling can be obtained.
[Brief description of the drawings]
[0076]
FIG. 1 is a diagram for explaining a relationship between a motor terminal voltage and a motor current in an H-bridge circuit.
FIG. 2 is a diagram for explaining improvement of the discontinuous characteristic of the motor terminal voltage to the motor current shown in FIG. 1 to a continuous characteristic;
3 is a diagram for explaining how the continuous characteristic of the duty ratio versus the voltage between the motor terminals shown in FIG. 2 is improved to a completely linear characteristic; FIG.
FIG. 4 is a diagram illustrating an outline of a configuration of an electric power steering apparatus.
FIG. 5 is a block diagram of an electronic control circuit.
FIG. 6 is a diagram illustrating an example of a configuration of a motor drive circuit.
FIG. 7 is a diagram illustrating a basic configuration of an H bridge circuit used as a motor drive circuit of an electric power steering apparatus.
FIG. 8 is a diagram for explaining a discontinuous portion that occurs in the relationship between the duty ratio of a motor current and a PWM signal.
FIG. 9 is a diagram for explaining a technique for solving a discontinuity that occurs in the relationship between the duty ratio of a motor current and a PWM signal.
[0077]
[Explanation of symbols]
  1 Steering handle
  2 axis
  3 Torque sensor
  4 Reduction gear
  5a, 5b Universal joint
  7 Pinion rack mechanism
  8 Tyrots
  9 Kratchi
  10 Motor
  11 Ignition Key
  12 Vehicle speed sensor
  13 Electronic control circuit
  14 Battery
  21 Phase compensator
  22A Steering assist command value calculator
  22B current controller
  23 Current Drive Linearizer
  24 Discontinuous current compensator
  25 Compensation adder
  26 multiplier
  27 Dead time compensator
  28 Duty Dither Adder
  30 Duty ratio calculation device
  35 Motor drive circuit
  36 FET gate drive circuit
  37 H Bridge circuit
  38 Motor current detection circuit

Claims (6)

In an electric power steering apparatus for controlling an output of a motor that applies a steering assist force to a steering mechanism based on a steering assist command value calculated based on at least a steering torque signal generated in a steering shaft,
A duty ratio calculating means for calculating a duty ratio D1 and a duty ratio D2 for determining a voltage between the motor terminals based on the steering assist command value;
A power source is connected between the input terminals of the H bridge circuit composed of the first and second arms each having two semiconductor elements connected in series, and the motor is connected between the output terminals. A motor driving circuit for driving an upper semiconductor element of the second arm with a PWM signal having the duty ratio D1, and driving a lower semiconductor element of the second arm with a PWM signal having the duty ratio D2.
When the duty ratio calculation means satisfies the following condition (c) between the absolute value Vref of the voltage command value between the motor terminals and the absolute value K _T ω of the back electromotive force of the motor ,
| Vref | <| K _T ω | (c)
The duty ratio D1 is calculated by the following formula (a), and the duty ratio D2 is calculated by the following formula (b).
D1 = Vref2 / V _r (a)
D2 = {Vref2 + sign (Vref2 ) (V r - | K T ω |)} / V r
・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ （B）
Vref: Motor terminal voltage command value
Vref2: Linearized motor terminal voltage command value
= 1/2 (Vref−K _T ω)
V _r : voltage supplied to the motor (battery voltage)
K _T : Motor back electromotive force constant
ω: Motor angular velocity
sign (Vref2): Electric power steering apparatus characterized by the sign of the linearized motor terminal voltage command value Vref2 . The duty ratio calculation means has the following condition (d) between the absolute value Vref of the motor terminal voltage command value from which noise is removed and the absolute value K _T ω of the back electromotive force of the motor from which noise is removed. When is satisfied
| Vref | <| K _T ω | (d)
The duty ratio D1 is calculated by the equation (a), and the duty ratio D2 is calculated by the equation (b).
The electric power steering apparatus according to claim 1 . When the duty ratio calculating means satisfies the following condition (f) between the absolute value Vref of the voltage command value between the motor terminals and the absolute value K _T ω of the back electromotive force of the motor ,
(| Vref | − | K _T ω |) <− Hys... (F)
However, Hys: Hysteresis width characteristic value
The duty ratio D1 is calculated by the equation (a) and the duty ratio D2 is calculated by the equation (b). When the following condition (g) is satisfied, the previous determination result is maintained.
−Hys <(| Vref | − | K _T ω |) <Hys... (G)
The electric power steering apparatus according to claim 1 . Noise removal processing is performed on at least one of the motor terminal voltage command value Vref and the motor back electromotive force K _T ω.
The electric power steering apparatus according to claim 3 . The hysteresis width characteristic value Hys is determined according to the magnitude of noise.
The electric power steering apparatus according to claim 3 or 4 , characterized in that: The duty ratio calculating means includes a current drive linearization compensator and a current discontinuity compensator, and the current drive linearization compensator is linear based on the equation (a) with the voltage command value Vref between motor terminals as an input. The duty ratio D1 corresponding to the normalized motor terminal voltage command value Vref2 is calculated, and the current discontinuity compensator calculates the duty ratio D2 based on the equation (b) with the linearized motor terminal voltage command value Vref2 as an input. thing
The electric power steering apparatus according to any one of claims 1 to 4 , wherein

Images